DE1254202B - Transformerless DC / DC converter - Google Patents

Description

FEDERAL REPUBLIC OF GERMANY

GERMAN

PATENT OFFICE

EDITORIAL

Int. Cl .:

H02m

German class: 21 a4 - 35/11

Number: 1254 202

File number: G37811IXd / 21 a4

Registration date: 61 V.

Display day: 3 #

The invention relates to a transformerless DC voltage converter with an inductance connected to a direct current source switch, which builds up the magnetic field of the inductance Electricity controls.

Many electrical utilities require an output voltage that is the same as that available Mains voltage exceeds. The usual method of getting a higher output voltage, is to step up within the voltage converter. With many Facilities, for example in power supplies in aircraft and spacecraft, in which a lightweight construction and fast regulation are of paramount importance is the use of a Transformer undesirable for this step-up transformation. Furthermore, with such a transformer the transmission ratio cannot be precisely adjusted, but to a finite number of steps limited. After all, the transformer cannot be used for AC and DC circuits use. You therefore need a DC / DC converter that does not require the use of a transformer increases the output voltage.

There are diode pulse amplifiers known, whose mode of operation is based on the storage of Charge carriers in the pn junction of a semiconductor diode is based. By complementing this well-known Amplifiers through low-pass filters can also be used to amplify continuous signals. Such diode amplifiers However, they belong to a different generic term, since the inductances of the low-pass filters do not serve to store energy and since a special driver source is also required.

The object on which the invention is based is to create such a DC voltage converter. This object is achieved by the invention specified in claim 1.

An expedient further development consists in designing the inductance as a tapped choke and to connect the switch to the tap point, so that the voltage at the switch is reduced.

Compared to the DC voltage converters according to the prior art, the one according to the invention is distinguished DC-DC converter is characterized by an extraordinary simplicity and is essential fewer components. The resulting low weight is of particular importance. Further be no electromechanical components required. The switching frequencies of the switch in the inventive DC-DC converters can be behavior-transformerless DC-DC converters

Applicant:

General Electric Company,
Schenectady, NY (V. St. A.)

Representative:

Dr.-Ing. W. Reichel, patent attorney,

Frankfurt / M. 1, Parkstrasse 13th

Named as inventor:

Burnice Doyle Bedford, Scotia, N.Y. (V. St. A.)

Claimed priority:

V. St. ν. America May 25, 1962 (197 626) - -

be chosen to be moderately high, which leads to a further spatial reduction. Becomes the DC / DC converter used in an AC powered system, which is often the case, so its design is independent of the frequency of the available alternating current, since the AC current is rectified before it is fed to the DC voltage converter.

In the following, the invention will be based on exemplary embodiments in conjunction with the drawings will be described in detail.

F i g. 1 shows a simplified schematic circuit diagram a DC-DC converter in accordance with the invention; the switching device is represented by a mechanical switch;

F i g. Figure 2 is a schematic diagram of the circuit of Figure 2. 1, one embodiment of the Switching device is shown in detail;

F i g. 3. shows the dependence of the output voltage on a control current at constant mains voltage, the entrance of the in F i g. 2 switching device shown is supplied;

Fig.4 is a schematic representation of the DC-DC converter of Fig. 1 showing another embodiment of the switching device;

5 shows a simplified schematic circuit diagram a DC / DC converter with which you can achieve a higher output voltage than with the circuit of FIG. 1.

The in F i g. 1 shown DC-DC converter has a pair of input terminals 1, 2, which with

709 688/137

a voltage source can be connected and the circuit supplies the input or supply voltage E _s. The output voltage E ₁ for the consumer 3 is taken from the output terminals 4, 5. A linear magnetic core coil 6, as used in filter circuits for power supplies, and a blocking diode 7 are in series between the input terminal 1 and the output terminal 4. The remaining terminals 2 and 5 are connected to one another and form the common reference point. A switching device 8, which is shown as a mechanical switch, has one terminal in common with the coil 6 and the diode 7. The other terminal is connected to the common reference point. A filter capacitor 9 is between the terminals 4 and 5 parallel to the consumer 3. Although in this simplified schematic representation the switching device is shown as a mechanical switch for better illustration, the switching device 8 is in reality a controlled circuit that the passage and blocking time of a switching element regulates as will be described later. The coil 6 and the switching device 8 together with the input terminals 1 and 2 form an input circuit which allows electrical energy from the voltage source to be stored in the coil 6 when the switching device 8 is closed. The switching device 8 and the diode 7 in series with it form, together with the consumer 3 and the capacitors 9 connected in parallel, an output circuit that feeds the stored electrical energy from the coil 6 to the capacitor 9 and the consumer 3 when the switching device 8 is open is. If the switching device 8 is opened and closed in rapid succession or put into the conductive and non-conductive state, then a smooth current flows from the input terminals 1 and 2 via the filter coil 6 and the filter capacitor 9 to the consumer 3.

In order to be able to understand the mode of operation of the DC / DC converter according to the invention, the prerequisites on which this mode of operation is based should be clarified once again. A necessary prerequisite is that the time constant L / R of the inductance 6 storing the electromagnetic energy and the time constant RC of the consumer circuit are large compared to the closing and opening times of the switch 8. Under these conditions, all current and voltage changes caused by the opening and closing of the switch can be regarded as linear changes without major errors, since in the series expansion of the changes which in and of themselves proceed according to an exponential function after the first, the linear link can break off. It is then possible to derive the voltage increase in the DC voltage converter according to the invention in the following way by simply considering the energy:

If the switch 8 is closed, electromagnetic energy is stored in the inductance 6. On the other hand, if the switch 8 is open, the inductance gives electromagnetic energy to the consumer 3 from. If the DC voltage converter has settled, it is in equilibrium, the amount of energy stored in the inductor when the switch is closed is equal to that when the switch is open Switch of the amount of energy delivered to the consumer.

The amount of energy β stored when the switch 8 is closed is calculated as follows

= E _s 7t _s

(1)

Here, E _{s means} the input voltage, that is to say the voltage across the inductance 6, 7 the mean current in the inductance and t _sc the time span during which the switch 8 is closed. It should be remembered that the average current 7 can be expected because the time constant of the inductance is so large that current changes during the period in which the switch 8 is closed are negligibly small.

The amount of energy Q that is delivered to the consumer during the period during which switch 8 is open is calculated as follows:

= (E _L -E _s ) according to _p

(2)

E _{L means} the voltage applied to the consumer and i _means the time span during which the switch 8 is open. It should be remembered once again that t _{s0 is also} small compared to the time constants of the inductance and in the consumer circuit.

Equating these two amounts of energy leads

on the following expression:

= (E _L -E _s ) lt _Si

(3)

After reducing it by 7 and dissolving it according to E _L , the following expression results:

E _L = E _s + E _s (t _sc / t _so ).

(4)

As can be seen from expression (4), the load voltage is only equal to the input voltage if the switch 8 always remains open. However, it is assumed here that the losses in the inductance are negligibly small. However, if the switch 8 is toggled, the ratio t _sc / t _{s0 becomes} greater than zero, and the load voltage increases above the input voltage.

The consumer _{voltage is therefore} dependent on the ratio t _sc / t s0 ". A circuit with which this ratio and thus the consumer voltage can be set will be described later.

It is advantageous if the closing time t of t _{sc sc sc} = 0 to i = t _so can be set. Then the load voltage E _{L changes} from E _L = E _s to Eι = 2E _s , as shown in FIG. 3 is shown. The switch 8 can, however, also be controlled in such a way that the ratio t _sc / t _so becomes greater than one, so that the load voltage still rises above 2E _s. In such a case, however, the losses occurring are greater than in conventional converter circuits with transformers, so that the circuits shown in FIGS. 2, 1 and 4 should generally only be used for those DC voltage converters in which the output voltages are not twice the input voltage exceed.

The losses in the components of FIG. 1 cause the output voltage E _{L to be} slightly smaller than was said before. The smallest inductance value of the coil 6 is determined by the switching device 8, since if the inductance is too small, the ripple of the current i _s becomes too great and prevents the switching device 8 from working properly.

The capacitor 9 reduces the ripple of the consumer current. The diode 7 causes the Capacitor 9 only discharges via the consumer circuit. The coil 6 serves to smooth the current and to increase the output voltage. The smoothed current has an effective decoupling between the input voltage source and the power amplifier, with all disturbances that from one circuit to the other can be reduced.

F i g. FIG. 2 shows an embodiment of the switching device 8 which can be implemented by various circuits. The requirements of such a circuit are that it _{is completely conductive or non-conductive during predetermined times i sc} and t and that the switching processes take place without contradiction. In addition, the switching device must be able to switch on and interrupt the line current i _s. A switching element which is conductive in one direction, the passage and blocking times of which can be precisely set, can be used to carry out the switching operations. A controlled silicon rectifier 10, which has a control electrode 11, a cathode and an anode, is well suited. A common silicon controlled rectifier is a pnpn semiconductor switch. It has similar properties to a gas-filled thyratron or ignitron, in which the control electrode loses its controllability as soon as the switching element is switched to the conductive state by a small control signal on the control electrode. In a recently developed rectifier, the control electrode retains its controllability.

The controlled rectifier 10 is by the signal of a control voltage source 12 in the conductive State shifted. The control pulses generated by 'the control voltage source 12 are with a Fixed repetition frequency of the control electrode 11 is supplied. The control voltage source 12 can be any of Square-wave or pulse generator exist, which is able to generate the control pulses with a constant Frequency, e.g. B. 2000 Hz, and the pulse length is sufficient to control the rectifier 10 to put in the conductive state. The control voltage source can also be a conventional unijunction transistor oscillator be.

The controlled rectifier 10 is blocked by a control circuit which is connected in parallel to the rectifier. The charging capacitor 13 and the saturable coil 14, which are connected in series, form part of the control circuit and are used to block the rectifier. The conduction time of the rectifier 10 is controlled by a circuit which has a saturable iron core transformer which has a primary winding 15 and a secondary winding 16. One end of the secondary winding is connected to the saturable coil 14 and the charging capacitor 13 via a blocking diode 17. The coil 14 and the transformer 15, 16 have essentially a rectangular hysteresis loop. A control current I _{c is} fed to the primary winding 15 of the transformer.

The control circuit has the effect that a blocking voltage is regularly applied from the capacitor 13 to the controlled rectifier 10 in order to reverse its current direction and block it. The operating period for the control current Ic- 0 is considered below. Shortly before the controlled rectifier 10 is opened by a pulse from the control voltage source 12, the potential of the charging capacitor is negative at its dotted end, and the coil 14 and the transformer 15, 16 are retentively negatively or positively saturated by the previous operating periods. As soon as the rectifier 10 is opened, the voltage of the capacitor 13 is applied to the coil 14 and to the secondary winding 16 of the transformer. In the process, a small magnetizing current flows which brings the coil 14 into positive saturation. As soon as the coil is positively saturated, the voltage on capacitor 13 reverses and tries to drive the coil into negative saturation. This is based on the oscillating ability of the capacitor 13 and the saturated inductance of the coil 14. When negative saturation is reached, the voltage of the capacitor is almost completely applied to the anode of the controlled rectifier 10, since the impedance of the saturated coil 14 is negligibly small. This voltage reverses the direction of the current in the rectifier 10. This blocks the rectifier. The conduction time of the controlled rectifier is determined by the time it takes to bring the coil 14 from negative to positive saturation and then back again to negative saturation. This period is precisely defined and is based on the parameters of the saturable coil.

The saturable transformer 15, 16 and the blocking diode 17. Leads used to control the conduction time of the controlled rectifier 10 to the primary winding 15 of the transformer to the maximum value of the control current I _c of about 150 mA to, then the secondary winding 16 of the transformer from the negative saturation brought almost to positive saturation, while coil 14 goes from negative to positive saturation, as described above. As soon as the coil 14 is positively saturated, the voltage on the capacitor 13 reverses its direction and has the effect that the coil 14 immediately tends towards negative saturation again, as described above. However, this sudden change in direction does not affect the secondary winding 16 of the transformer, since the blocking diode 17 prevents a current direction reversal. The secondary winding 16 must be brought back into negative saturation by the control current I _c.

F i g. 3 shows the dependence of the output voltage E _L on the control current I _c in FIG. 2 power amplifier shown. The characteristic curve shows that when the control current I _{c is} reduced to some intermediate value, e.g. B. 20 mA, the circuit works on the falling part of the characteristic and the output voltage E _L - is reduced proportionally. The reduction in the output voltage is due to the fact that the smaller control current which is fed to the primary winding 15 of the transformer can no longer bring the secondary winding 16 to negative saturation before the control voltage source 12 initiates the next operating period. A consequence of this is that as soon as the controlled rectifier 10 is put into the conductive state by the control voltage source 12, the charging current brings the secondary winding 16 into positive saturation much faster and the capacitor is already uncharged before the coil reaches positive saturation -

enough. The reversal of the voltage on the capacitor 13 immediately causes a current change in the coil 14, so that the coil returns to negative saturation. A smaller hysteresis loop is run through. As soon as the coil 14 is negatively saturated, the voltage of the capacitor 13 is applied to the controlled rectifier 10. The direction of the current is reversed and the rectifier is blocked, as described above. The rectifier is now blocked earlier because the coil 14 only runs through a small hysteresis loop. The shorter passage time (smaller ratio t _sc / t _S o) has the effect that less electrical energy is stored in the coil 6. As a result, less electrical energy is supplied to the consumer 3 and the output voltage E _{L is} reduced to approximately 200 V. The characteristic curve shows that this voltage corresponds to a control current of 20 mA. The saturable transformer 15, 16 thus controls the length of time that the coil 14 needs to reach negative saturation. During this time the controlled rectifier is blocked.

The in F i g. The invention shown in FIG. 4 shows another circuit which can be used for the switching device 8 in FIG. 1 can use. In the case of the in FIG. 4, the passage time of the controlled rectifier 10 is fixed and the switching frequency is adjustable. The control voltage source 12 is shown in FIG. 4 is shown as a control circuit, while the circuit that blocks the rectifier is made up of solid components, namely the capacitor 13 and the saturable coil 14. The circuit of FIG. 4 therefore has a controllable unijunction transistor frequency oscillator. The fixed conduction time of the rectifier is determined by the capacitor 13 and the coil 14. The advantage of this circuit is that you can still work at very low frequencies with the help of the variable frequency of the transistor oscillator. The regulation of the output voltage E _L is more precise, in particular in the case of ratios EJE ₅ which approach one. This precise regulation in the case of small ratios _{EJE 5} can be achieved with the control circuit from FIG. 2, since it takes a certain time to block the rectifier and the ratio t _sc / t _can not be made sufficiently small.

The variable control voltage source 12 has a unijunction transistor 18. It works like a conventional pulse oscillator, the control signals for the control electrode 11 of the controlled rectifier 10 supplies. The rectifier is opened at predetermined times that are determined by the frequency of the Unijunction transistor oscillator are set. The frequency is determined by the resistor 19 and the capacitor 20 connected in series. The connection point of these two components is connected to the emitter of the unijunction transistor 18. The series connection with the transistor 21, the collector of which is connected to the other end of the resistor 19, and to the Diode 22 connected to the emitter of transistor 21 determines the one previously set Current consumption of the capacitor 20, the frequency of the unijunction transistor oscillator being exactly is determined. With this precise control, the oscillator frequency can be set from 0 to 3 kHz will. A considerable control range of the output voltage compared to the supply voltage is obtained.

The transistor 23, which is in series with the resistor 24 and whose collector is connected to the base of the transistor 21, works as a potentiometer and defines the operating point of the transistor 21. A control voltage E _c is applied to the base of the transistor 23 via a current limiting resistor 25. This control voltage represents the operating point of the transistor 23, which then determines the operating point of the transistor 21, whereby the frequency of the unijunction transistor oscillator circuit is determined. When the control voltage E _C = Q , then the transistors 23 and 21 are blocked and the frequency of the transistor oscillator circuit is zero. The output voltage E _L is then equal to the supply voltage E ₅ . The diode 22 has the effect that the transistor 21 is always blocked at the control voltage Eq = Q , regardless of any temperature changes within the circuit. If the control voltage E _{c increases} in the positive direction, then the transistors 23 and 21 become more and more conductive, and the frequency of the unijunction transistor oscillator increases. In this way, the control voltage E _c regulates the frequency of the oscillator, which in turn controls the output voltage E _1. A voltage divider with resistors 26 and 27, which is located between input terminals 1 and 2, supplies the transistor circuit with the necessary supply voltage. The resistor 28 is connected to one base terminal of the unijunction transistor 18 and limits its base current. The output voltage of the oscillator drops across the resistor 29, which is connected to the other base terminal of the unijunction transistor 18 and to the control electrode 11 of the rectifier 10.

F i g. 5 shows a modification of the simplified circuit of FIG. 1. With the modified circuit, an output voltage E _{L can be} obtained that is significantly greater than the supply voltage E _s . Although the circuit of Fig. 1 can produce output voltages that are more than twice as large as the supply voltage, this circuit is not suitable for such high voltages, since the switching device 8 has the entire output voltage and the maximum peak voltages of the semiconductor components can be exceeded. The voltage on the switching device 8 can be reduced by tapping on the coil 6. The output voltage can then be a multiple of the supply voltage. A coil with a tap makes it very easy to produce an output voltage of 750 V with an input voltage of 125 V.

Claims (4)

Patent claims: 1. Transformerless DC voltage converter with a switch connected to an inductance to a direct current source and which controls the current building up the magnetic field of the inductance, characterized in that the switch (8) contains a controllable rectifier (10) which is controlled by a control device (12) at predetermined time intervals ) can be switched on and switched off by a switching device (13 to 17), and that the time constant LIR of the in- ductility (6) and the time constant RC of the consumer (3) are large compared to the switching times of the switch (8). 2. DC voltage converter according to claim 1, characterized in that the inductance (6) is designed as a tapped throttle (FIG. 5) and the switch (8) is connected to the tapping point is, so that the peak voltage at the switch (8) is reduced. 3. DC voltage converter according to claim 1 or 2, characterized in that a smoothing capacitor (9) in parallel with the consumer (3) and a rectifier (7) in series between the Inductance (6) and the consumer is switched. IO 4. DC voltage converter according to one of claims 1 to 3, characterized in that the switching device (13 to 17) is a saturable, in series between the inductance (6) and contains the switching capacitor (13) switched choke (14), the center tap with the controllable rectifier (10) is connected (Fig. 4). Considered publications: "IRE Transactions on Electron Devices", 1959, pp. 341 to 346. Legacy Patents Considered:
German Patent No. 1 203 838. 1 sheet of drawings 709 688/137 11.67 © Bundesdruckerei Berlin